Vision based behavior analysis of laboratory animals: A 3D gradient based approach

In pharmacological experiments behavior pattern of laboratory mice, which are under the influence of psychotherapeutic drugs, reveals important clues about efffects of the drug. Behavior analysis of laboratory mice by video processing saves both time and labor. In this work a method which was previously used to recognize human behaviors is adapted to laboratory mice case. Method is based on fitting histograms of spatio-temporal gradients extracted from 3D space-time volumes to multidimensional statistical distributions and classifying according to distances between the distributions. In this work the method is tested on a common mice video dataset, compared to other methods in the literature and found to be successful.